The connection of monocytes and reactive oxygen species in pain

Abstract

The interplay of specific leukocyte subpopulations, resident cells and proalgesic mediators results in pain in inflammation. Proalgesic mediators like reactive oxygen species (ROS) and downstream products elicit pain by stimulation of transient receptor potential (TRP) channels. The contribution of leukocyte subpopulations however is less clear. Local injection of neutrophilic chemokines elicits neutrophil recruitment but no hyperalgesia in rats. In meta-analyses the monocytic chemoattractant, CCL2 (monocyte chemoattractant protein-1; MCP-1), was identified as an important factor in the pathophysiology of human and animal pain. In this study, intraplantar injection of CCL2 elicited thermal and mechanical pain in Wistar but not in Dark Agouti (DA) rats, which lack p47(phox), a part of the NADPH oxidase complex. Inflammatory hyperalgesia after complete Freund's adjuvant (CFA) as well as capsaicin-induced hyperalgesia and capsaicin-induced current flow in dorsal root ganglion neurons in DA were comparable to Wistar rats. Macrophages from DA expressed lower levels of CCR2 and thereby migrated less towards CCL2 and formed limited amounts of ROS in vitro and 4-hydroxynonenal (4-HNE) in the tissue in response to CCL2 compared to Wistar rats. Local adoptive transfer of peritoneal macrophages from Wistar but not from DA rats reconstituted CCL2-triggered hyperalgesia in leukocyte-depleted DA and Wistar rats. A pharmacological stimulator of ROS production (phytol) restored CCL2-induced hyperalgesia in vivo in DA rats. In Wistar rats, CCL2-induced hyperalgesia was completely blocked by superoxide dismutase (SOD), catalase or tempol. Likewise, inhibition of NADPH oxidase by apocynin reduced CCL2-elicited hyperalgesia but not CFA-induced inflammatory hyperalgesia. In summary, we provide a link between CCL2, CCR2 expression on macrophages, NADPH oxidase, ROS and the development CCL2-triggered hyperalgesia, which is different from CFA-induced hyperalgesia. The study further supports the impact of CCL2 and ROS as potential targets in pain therapy.

Figure 1. CCL2-induced hyperalgesia in Wistar, but not in Dark Agouti (DA) rats.

Mechanical (A, B) and thermal (C, D) nociceptive thresholds were quantified before and after i.pl. injection of CCL2 (0.3 µg – open circle, 1 µg – filled triangle, 3 µg – open triangle) or 0.9 % NaCl (solvent  =  control – filled circle) into Wistar (A, C) and DA (B, D) rats (*p<0.05 versus time point 0 h, Two Way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM of raw values (A, B) or % maximal possible effects (MPE) (C, D).

Figure 2. Comparable nociceptive thresholds and DRG currents in Wistar and Dark Agouti (DA) rats.

Rats received an i.pl. injection of CFA (A–D; ipsilateral injected paw  =  open circle, contralateral, non injected paw  =  filled circle). Mechanical (A, B) and thermal nociceptive (C, D,) thresholds were quantified (*p<0.05 versus time point 0 h, Two Way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM of raw values (A, B) or % maximal possible effects (MPE) (C, D). (E) DRG-neurons from naïve DA rats were obtained and cultured for 48 h. Representative traces of whole cell recordings are shown after application of capsaicin (500, 1000 nM, left). Bar graph quantifies capsaicin-activated inward currents in DRG neurons from Wistar and DA rats (right) (n = 9, means ± SEM, p>0.05, t- test). In addition DA rats received i.pl. injection of capsaicin (F; solvent – filled circle, 30 µg capsaicin – open circle) and thermal nociceptive thresholds were quantified (*p<0.05 versus time point 0 h, Two Way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM of % maximal possible effects (MPE).

Figure 3. Defective CCL2-induced chemotaxis of macrophages from Dark Agouti (DA) in comparison to Wistar rats.

Wistar (A) and DA (B) rats were i.pl. injected with 3 µg CCL2 (light grey bars), with CFA (dark grey bars, positive control) or with 0.9% NaCl (black bars, solvent negative control). Single cell suspensions were prepared from paw tissue and the number of infiltrating ED1⁺ CD45⁺ macrophages was quantified by flow cytometry (n = 4, *p<0.05, One Way ANOVA, Student-Newman-Keuls). (C) CCR2 mRNA from peritoneal rat macrophages from Wistar (white bars) as well as DA (black bars) rats was measured with real time PCR (*p<0.05, t-test, n = 6). (D) The migratory capacity of macrophages from Wistar rats (white bars, n = 8) and DA rats (black bars, n = 5) was quantified in a Boyden chamber in response to CCL2 and solvent control (p<0.05, One way ANOVA, Dunns-Method).

Figure 4. Reconstitution of CCL2-induced hyperalgesia by adoptive transfer of macrophages derived from Wistar, but not from Dark Agouti (DA) rats.

(A) Mechanical nociceptive thresholds were quantified after i.pl. injection of 3 µg CCL2 with prior leukocyte depletion by systemic injection of CTX (filled triangles), 3 µg CCL2 with prior systemic injection of solvent (cyclo-phosphamid (CTX) treatment (open circles), or solvent of CCL2 without CTX treatment (filled circles). CTX injections were performed 3 d with 10 mg/kg and 1d 50 mg/kg before cell and/or CCL2 treatment. (B) Representative dot blots (x-axis: forward scatter (FSC) – cell size; y-axis: sideward scatter (SSC) – cell granularity) for leukocyte subpopulations in peripheral blood of untreated (left) and CTX-treated Wistar rats (right) are shown by flow cytometry. Gates were set on beads (for quantification), neutrophils, lymphocytes and monocytes. (C–F) Wistar and DA rats were leukocyte-depleted by i.p. injection with CTX followed by i.pl. injection of different numbers of macrophages from either Wistar or DA rats, 30 min later, by i.pl. injection of 3 µg CCL2. Injection of 2×10⁶ macrophages without concomitant injection of CCL2 served as a negative control. (C) Wistar rats reconstituted with cells from Wistar rats. I.p. injections of solvent only served as a negative control for CTX treatment (open triangles) with the same injection pattern. (D) Wistar rats reconstituted with cells from DA rats. (E) DA rats reconstituted with cells from Wistar rats. (F) DA rats reconstituted with cells from DA rats. (*p<0.05 compared to time point 0 h, Two Way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM.

Figure 5. Establishment of CCL2-induced hyperalgesia following restoration of defective ROS generation in Dark Agouti (DA) rats.

Peritoneal macrophages from Wistar and DA rats were isolated and treated for 1 h with 0.2 µg CCL2 or solvent control. ROS production was quantified by flow cytometry using the phagoburst assay. Representative histograms are shown for Wistar rats (A) and DA rats (B) with solvent (grey), CCL2 (black line) or positive assay control PMA (dotted line). (C) Results were analyzed statistically (n = 6/group, CCL2 i.pl. (white bars) or solvent (black bars), *p<0.05, t-test). (D) Wistar and DA rats were i.pl. injected with CCL2 i.pl. (lower panel) or solvent (upper panels) for 3 h, subcutaneous paw tissue was representative immunohistochemistry is shown for HNE (red), ED1⁺ macrophages (green) and merge of both together with nuclei (DAPI-blue). (E) Wistar and DA rats were i.pl. injected with CCL2 i.pl. (white bars) or solvent (black bars) for 3 h, subcutaneous paw tissue was prepared and HNE adducts were quantified by ELISA (n = 6/group, *p<0.05, t-test). (F) DA rats were s.c. injected with phytol for 5 days. Mechanical nociceptive thresholds were quantified before and after i.pl. injection of CCL2 (0.3 µg – open circle, 1 µg – filled triangle, 3 µg – open triangle). Control animals were treated 5 days with 0.9 % NaCl before i.pl. injection of 3 µg CCL2 (filled circle), (*p<0.05 versus time point 0 h, Two way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM.

Figure 6. Inhibition of CCL2-induced mechanical hyperalgesia by ROS scavenging in Wistar rats.

Mechanical nociceptive thresholds were quantified before and after (A) i.pl. injection of 1 µg apocynin with 3 µg CCL2 (filled circles) or 3 µg CCL2 alone and (B) i.pl. injection of 1 µg apocynin (filled circles) or its solvent 0.9 % NaCl (open circles) in rats with 4 d CFA inflammation. (C–E) Mechanical nociceptive thresholds were quantified before and after i.pl. injection of 3 µg CCL2 alone with the respective solvent (filled circle) or together with (C) SOD (3 U open circle, 30 U filled triangle, 300 U open triangle), (D) catalase (2–5 U – open circle, 20–50 U – filled triangle, 200–500 U – open triangle), or (E) tempol i.p. (1.5 mg – open circle, 7.5 mg – filled triangle, 15 mg – open triangle) (*p<0.05 versus time point 0 h, Two Way RM ANOVA, Student-Newman-Keuls, n = 6). Data are presented as means ± SEM.